Singapore Academic Cybersecurity R&D
Harnessing R&D to Secure our Nation
AppMod: Safety and Privacy of Smart-City Mobile Applications through Model Inference
- Lead PIs : David Lo, Associate Professor, SMU ( email@example.com ) and Eran Toch, Professor, Tel Aviv University ( firstname.lastname@example.org )
- Host Institution : School of Information Systems, SMU
- Partner Institution : Tel Aviv University
Increasing mobile platforms reliability by detecting anomalies and allowing users to effectively respond to them
WP1: Generation of rich fine-grained logs and stateful model inference
We have developed a prototype that can modify an Android apk to generate rich logs of sensitive system calls. We have also developed a powerful model inference algorithm that performs deep learning for mining finite state automaton (FSA)-based specifications. A paper describing the latter work has been reviewed by NRF and approved for publication.
WP2: Differencing analysis and anomaly detection
We have developed a new approach to generate succinct summaries that characterize the differences between two logs or within a set of many logs. The differences are presented using a concise finite-state model that describes and highlights similarities and differences among the logs.
WP3: Crowd-centric, socially-aware user interaction model
We have built a preliminary version of a crowd-centric, demographic aware mobile app where users can get help from and provide help to others when malicious/questionable activities (e.g., an app is trying to send contact information sent over the internet) are occurring in their mobile phones. We have also conducted a survey to investigate what kind of help did people ask from and give to others when operating their mobile phones and configuring security and privacy settings.